An enantiomeric interaction in the metabolism and tumorigenicity of (+)- and (-)-benzo[a]pyrene 7,8-oxide.

The (+)- and (-)-enantiomers of benzo[a]pyrene 7,8-oxide are hydrated stereospecifically at C-8 to (-)- and (+)-trans-7,8-dihydroxy-7,8-dihydrobenzo[a]pyrene, respectively, by rat hepatic epoxide hydrolase. The (-)-enantiomer of benzo[a]pyrene 7,8-oxide is metabolized by microsomal epoxide hydrolase at a rate 3- to 4-fold greater than the (+)-enantiomer. At low conversion of racemic substrate, however, benzo[a]pyrene 7,8-oxide is metabolized to the dihydrodiol at a rate equal to that of the (+)-enantiomer. An analysis of the enantiomeric composition of the dihydrodiol formed from the racemic substrate revealed preferential formation of (-)-trans-7,8-dihydroxy-7,8-dihydrobenzo-[a]pyrene. At low substrate conversion (< 20% metabolism), the enantiomeric purity of the dihydrodiol was much higher than at high substrate conversion (> 50% metabolism). Similar results were obtained with microsomes from hamster, rabbit, guinea pig, mouse, and human liver. These results indicate that epoxide hydrolase has a higher affinity for (+)-benzo[a]pyrene 7,8-oxide than for the (-)-enantiomer. The kinetics of hydration of (+)- and (-)-benzo[a]pyrene 7,8-oxide by purified epoxide hydrolase in detergent solution showed the (+)- and (-)-enantiomers to have apparent Km values of 1.7 and greater than or equal to 20 microM, respectively. Tumorigenicity studies with benzo[a]pyrene 7,8-oxide on mouse skin and in newborn mice revealed that (+)-benzo[a]pyrene 7,8-oxide, the metabolic precursor of the more tumorigenic (-)-7,8-dihydrodiol, is significantly more tumorigenic than the (-)-enantiomer. However, racemic benzo[a]pyrene 7,8-oxide was more tumorigenic than either enantiomer alone, indicating an enantiomeric synergism in the carcinogenicity of benzo[a]pyrene 7,8-oxide. The data are discussed in relation to the complete sequence of metabolic pathways leading to an ultimate carcinogen from benzo[a]pyrene.

in the mutagenic and carcinogenic potency of enantiomeric metabolites of polycyclic aromatic hydrocarbons (1). Since these metabolites are formed by enzymes which show varying degrees of stereoselectivity, the role of stereochemical factors in the metabolism of a carcinogen may be of critical importance in the carcinogenic process. In this regard, the stereoselective' metabolism of BP2 to an ultimate carcinogen has been well documented. For the metabolic sequence B P -+ B P 7,8-oxide -+ BP 7,8-dihydrodiol -B P 7,8-diol-9,10-epoxide, only one of the possible enantiomers of the dihydrodiol and diol epoxide displays high carcinogenic activity (1,3-6). However, no information is present.ly available on the first step of this metabolic sequence, i.e. the enantiomeric purity and biological activity of the metabolically formed BP 7,8-oxide. Since the microsomal mixed function oxidase system and epoxide hydrolase function sequentially to form (-)-BP 7,8dihydrodiol of high enantiomeric purity" (>90%) from BP, either or both of these enzymes play a role in determining the enantiomeric composition of the metabolically formed dihydrodiol. Conflicting results have been obtained on the stereoselective metabolism of racemic B P 7.8-oxide to the dihydrodiol. Thakker et at. (7,8) have reported that racemic 7,8-oxide is metabolized by epoxide hydrolase to (-)-BP 7,8-dihydrodiol of very low enantiomeric purity (7 to 10%) while Yang et al. (9,10) have reported high enantiomeric purity for (-)-BP 7,8dihydrodiol (72%). The recent synthesis (11) and assignment of absolute stereochemistry (12) of the enantiomers of BP 7.8oxide by x-ray crystallography ( Fig. 1) has permitted us to reexamine the stereoselective metabolic factors involved in the formation and metabolism of BP 7,8-oxide to BP 7,8-dihydrodiol. In addition, the carcinogenic activity of the enantiomers of BP 7,8-oxide have been examined to determine the complete sequence of metabolic events leading to an ultimate carcinogen of BP.
The Enzyme Assays-Immature male rats (50 to 60 g) of the Long-Evans strain (Blue Spruce Farms, Altamont, N. Y.) were treated with phenobarbital (75 mg/kg/day, intraperitoneally) or 3-methylcholanthrene (25 mg/kg/day, intraperitoneally) for 4 days prior to being killed. Microsomes from control, 3-methylcholanthrene-and phenobarbital-pretreated rats were prepared as previously described (16) and were stored at -90°C. Epoxide hydrolase was purified to apparent homogeneity as described (17). Metabolism of [B-"H]-(+.)-BP 7,8-oxide (8.5 pCi/pmol) and [3-"H]ously described (18). Rates of hydration of unlabeled (+)-and (-1-(-+)-phenanthrene 9,lO-oxide (7.4 pCi/pmol) was measured as previ-BP 7,8-oxide to B P 7.8-dihydrodiol by microsomal epoxide hydrolase were measured by a modification of the enzyme assay with radioactive substrates (18). The reaction mixture contained 25 pl of 0.5 M Tris-HC1 buffer (pH 8.9 a t 37"C), 50 pl of water and enzyme solution, and 5 pl of acetonitrile:NH40H ( l w l ) containing 15 nmol of BP 7,8oxide. Samples were incubated at 37'C for 2 min and the reaction was stopped by the addition of 25 pl of tetrahydrofuran. One-third of the total mixture ( Kinetics of the hydration of (+)-and (-)-BP 7,8-oxide by purified epoxide hydrolase were followed at 381 nm on a Beckman Acta V spectrophotometer thermnstated at 25'C. Reactions were carried out in 3.0 ml of 50 mM Tris, 100 mM NaCl (pH 8.5 a t 25°C) containing 20 mM Brij 35. Substrates were added in 10,111 of tetrahydrofuran:NH,OH (1oOO:l) to a final concentration of 6 to 8 p~. Reactions were initiated by addition of 25 to 50 p1 of purified epoxide hydrolase solution. The catalytic rate constant, kc, and apparent K , (K,,,,) were obtained by fitting the time course data to the appropriate integrated rate expression.
Determination of the Enantiomeric Purity of Metabolically Formed BP 7,8-Dihydrodiol-BP 7,8-oxide, 800 nmol in 40 pI of tetrahydrofuran:NH40H (lOOO:l), was incubated with 2.5 mg of microsomal protein from phenobarbital-pretreated rats and 0.30 mmol of Tris-HC1 buffer (pH 8.9 at 37OC) in a final volume of 2.0 ml for 1 or 10 min at 37°C. Reactions were terminated by the addition of 6 ml of ethyl acetate. The samples were blended on a Vortex mixer for 1 min and centrifuged. The organic phase was removed, dried with 5 g of sodium sulfate, and evaporated to dryness under N? after addition of an ethyl acetate wash of the sodium sulfate. Separation of B P 7,8dihydrodiol from the incubation mixture, as well as formation and analytical separation of its bis-MTPA diesters by high pressure liquid chromatography were as previously described (7,8). Extent of dihydrodlol formation was quantified by absorbance at 367 nm in methanol (E = 50,500 M" cm") or radiochemically, and enantiomeric purity was established from the areas of the diastereomeric bis-MTPA peaks upon high pressure liquid chromatography as well as radiochemically when possible. Me2SO:NH40H (1ooO:l) a t a concentration of 100 nmol/5 pl. The mice were given a total dose of 700 nmol of B P 7,s-oxide divided into three injections of 100, 200, and 400 nmol administered on the lst, 8th, and 15th day of life, respectively. Control mice were injected with solvent. The mice were weaned a t 23 days of age, and the experiment was terminated by killing the animals when they were 31 to 35 weeks of age. At autopsy, the major organs of each animal were examined grossly, tumors were counted, and tissues were fixed in 10% buffered formalin. A representative number of pulmonary tumors, all hepatic tumors, and all other tissues with suspected pathology were examined histologically, Pathology of the lung tumors was the same as has been previously described (19). Liver tumors were characterized as neoplastic nodules (20).

Metabolism of (+)-and (-)-BP 7,8-Oxide by Microsomal
Epoxide Hydrolase-Formation of BP 7,8-dihydrodiol from (+)-and (-)-BP 7,8-oxide by rat hepatic microsomal epoxide hydrolase was proportional to enzyme concentration (Fig. 2). The (-)-enantiomer was metabolized at a rate approximately 4 times greater than the (+)-enantiomer. When racemic 7,8oxide was incubated with hepatic microsomes however, metabolism occurred at a rate equal to the (+)-enantiomer when less than 20% of the substrate was converted to product. The rate of metabolism of unlabeled racemic BP 7,8-oxide (6.3 nmol/min/mg of protein) with the newly developed fluorometric assay was comparable to values previously obtained with a radiometric assay (18). Table I  were incubated with 375 PM [3-3H]phenanthrene 9,lO-oxide (left) or 125 p~ [3-"Hlphenanthrene 9.10-oxide (right) in the presence of the PM concentration. This same percentage inhibition of phenanthrene 9,lO-oxide hydration was obtained with 15 p~ (+)-BP 7,8-oxide, indicating at least a 12-fold difference in the ability of the two enantiomers to inhibit the hydration of phenanthrene 9,10-oxide.
Hydration of BP 7,8-oxide by purified epoxide hydrolase can be followed spectrophotometrically at 381 nm (A6 = -13,000 M" cm") as shown in Fig. 4. Complications from the competing spontaneous isomerization of the substrate were avoided by monitoring the reaction at an isosbestic point for the isomerization (381 nm) and by the use of high enzyme concentrations. BP 7,8-oxide isomerizes spontaneously to phenol (monitored at 354 nm) with a half-life of 52.7 min under the conditions employed. Thus, the spontaneous isomerization occurs 10 times more slowly than the slowest enzyme-catalyzed reaction, resulting in 295% recovery of BP ?&dihydrodiol based on starting oxide. The enzyme-catalyzed hydration of the (+)-enantiomer in detergent solution followed Michaelis-Menten kinetics with a Kmapp = 1.7 p~ and k,. = 0.10 s" (Fig. 4A). In contrast, the hydration of the (-)enantiomer followed first order kinetics under the conditions employed, suggesting that K,,,, >> [SI (Fig. 4 B ) . This kinetic behavior allows a lower limit of KmaPr, for the (-)-enantiomer to be set at 220 p~. Furthermore, a ratio of the second order  rate constants (kc/Kmapp) for the two enantiomers (kc/KmaPp)+/ (ke/Kmapp)-= 15 can be calculated. Since the catalytic rate constant for the hydration of (-)-BP 7,8-oxide is greater than that for (+)-BP 7,8-oxide (Fig. 2), the ICmapp for the (-)enantiomer must be >15 times the ICmapp for the (+)-enantiomer. Thus, the kinetic results obtained with purified epoxide hydrolase in detergent solution, which have previously been proposed to mimic microsomal kinetic conditions (21), are consistent with the above results obtained with microsomal suspensions.
Since hepatic epoxide hydrolase from rabbit, hamster, guinea pig, and human have different antigenic properties than the rat enzyme (23), we investigated the metabolism of racemic BP 7,g-oxide to BP 7,8-dihydrodiol by various species. Mouse liver epoxide hydrolase, which is antigenically identical to the rat enzyme (23), as well as enzyme from hamster, rabbit, guinea pig, and human, all formed (-)-BP 7,8-dihydrodiol of high enantiomeric purity when less than 30% of the substrate was consumed in the reaction (Table 11)

-oxLde in newborn mice
Swiss-Webster mice (80 animals/group) were given intraperitoneal injections of 100 nmol, 200 nmol, and 400 nmol of BP 7,8-oxide on the lst, 8th, and 15th day of life, respectively. The animals were weaned at 25 days of age and the experiment was terminated when the animals were 31 to 35 weeks of age. However, in both tumor models, racemic BP 7,8-oxide produced a higher tumor incidence than either optical enantiomer. Without any enantiomeric interaction, the tumorigenic activity of racemic BP 7,8-oxide should have been intermediate to the activity of the (+)-and (-)-enantiomers. This is not the case, and the above results indicate that there is a synergistic interaction in the tumorigenicity of the optical enantiomers of BP 7,8-oxide.

DISCUSSION
Although stereochemical factors have long been known to be involved in the metabolism and expression of biological The reader should note that the assignment of absolute configuration at C-9 in the present paper is indeed correct. In our original publication (29), this center was inadvertently referred to as 9 s when it should have been 9R and 9R when it should have been 9s. The structures in that study and in the present study are correct as shown. activity of isomeric molecules, only recently has the importance of these factors been demonstrated in the expression of mutagenicity and carcinogenicity. For the well studied polycyclic aromatic hydrocarbon BP, marked differences in the metabolism, mutagenicity and carcinogenicity of the biologically active metabolites has been demonstrated (1). Previous studies (7)(8)(9)(10)27,28), together with the results presented here, have now established the stereochemical factors involved in the sequence of reactions for the metabolism of BP to an ultimate carcinogen (Fig. 7). The results presented here demonstrate that (+)-and (-)-BP 7,8-oxide are stereospecifically metabolized to (-)-and (+)-BP 7,8-dihydrodiol, respectively, by hepatic microsomal epoxide hydrolase. While the rat enzyme metabolizes the (-)-enantiomer at a rate 3 to 4 times greater than the (+)-enantiomer, epoxide hydrolase has a higher affinity for the (+)-enantiomer compared to the (-)enantiomer. Thus, when various ratios of the two enantiomers are incubated, the (+)-enantiomer is preferentially metabolized. At low conversion of substrate, this results in the formation of (-)-BP 7,8-dihydrodiol of high enantiomeric purity. Extensive metabolism of racemic B P 7,8-oxide results in the formation of large amounts of both enantiomers of the dihy-drodiol. These results probably explain the apparent discrepancy in the reported enantiomeric purity of BP 7,8-dihydrodiol formed from racemic BP 7,8-oxide. Thakker et al. (7,8 ) reported low enantiomeric purity at high substrate conver-sion5 while Yang et al. (9,10) reported high enantiomeric purity at low conversion (520%) of substrate.
Metabolism of BP to BP 7,8-dihydrodiol via the sequential action of the mixed function oxidase system and epoxide hydrolase results in the formation of predominantly the (-)enantiomer The high enantiomeric purity of the dihydrodiol could be the result of one of the two following metabolic events: ( i ) formation of highly enantiomerically pure BP 7,8-oxide or ( i i ) formation of BP 7,8-oxide of low enantiomeric purity followed by preferred hydration of (+)-BP 7,8-oxide at low substrate conversion (see "Results").
Thus, to simulate the conditions under which BP 7,8-dihydrodiol is formed from the parent hydrocarbon, we have performed experiments in which racemic BP 7,8-oxide was infused, with and without the addition of other metabolically formed BP arene oxides, into an incubation with hepatic microsomes at a rate equal to the formation of the corresponding dihydrodiols from BP (32). In these incubations, almost equal amounts of (+)-and (-)-BP 7,8-dihydrodiol were formed even in the presence of other competing BP arene oxide substrates (Table V). However, when BP is incubated under these conditions, greater than 9 0 % of the BP 7,8-dihydrodiol formed, due to the sequential action of the mixed function oxidase system and epoxide hydrolase, is the (-)enantiomer (1, 7). These results provide the only unequivocal evidence for the metabolism of BP to (+)-BP 7,8-oxide of high enantiomeric purity by rat liver microsomes. It appears, however, that the metabolic formation of (+)-BP 7,8-oxide is stereoselective and not stereospecific since small amounts of (+)-BP 7,8-dihydrodiol (4 to 8% of the total BP 7,8-dihydrodiol formed) are detected when BP is metabolized by rat liver microsomes.6 In contrast to the conclusion by Yang et ul. (9,10) that a single enantiomer of BP 7,8-oxide is formed by rat liver microsomes, the present results indicate that both enantiomers are formed albeit the (+)-enantiomer is formed in 10-to 20-fold excess. In addition, when BP is applied topically to mouse skin (33) or to cells in culture (34)(35)(36)(37), enantiomers of the diastereomeric BP 7,8-diol-9,10-epoxides derived from both (+)-and (-)-BP 7,8-dihydrodiol are detected as adducts bound to macromolecules, although the covalently bound product(s) derived from the (-)-enantiomer predominate. Fig. 7 summarizes the metabolic pathways involved in the stereoselective metabolism of BP to the bay region 7,8-diol-9,lO-epoxides. It is of interest to note that the enzymes made all of the wrong stereochemical choices in their conversion of BP to the BP 7,8-diol-9,10-epoxides as f a r as the welfare of the animal was concerned. (+)-7/3,8a-Dihydroxy-9a,lOaepoxy-7,8,9,10-tetrahydrobenzo[a]pyrene [ (+)-BP-7,8-diol-9,10-epoxide-2] is the most potent carcinogenic diol epoxide of BP (5,6), and it is formed from BP to a considerably greater extent than are the other three diol epoxide isomers (7). If the mixed function oxidase system had metabolized BP exclusively to (-)-BP 7,8-oxide, the resulting diol epoxides would have had very markedly reduced carcinogenicity.
An evaluation of the tumorigenic activity of the enantiomers ' Results presented here indicate that per cent conversion of (&)-B P 7,8-oxide were underestimated in our previous studies (7, 8).
Yang and co-workers (9, 10, 30, 31) have indicated that the BP 7,8-dihydrodiol formed from B P by rat liver microsomes is enantiome r i c d y pure. However, their own data (9,31) indicates that detectable amounts of the (+)-enantiomer are formed, which is consistent with the prior report of Thakker et al. (7) and further confirmed by Jerina et al. (1). Interestingly, an enantiomeric enhancement in the tumorigenicity of racemic BP 7,8-oxide was also observed in these studies. These results are the first demonstration of a synergistic carcinogenic effect for enantiomers. Although the reason(s) for this synergism are not known, metabolic factors described above could contribute to this phenomenon. It should be emphasized, however, that factors other than metabolism may be involved in the higher than expected tumorigenic activity of racemic BP 7,8-oxide. We recently reported (38) a synergistic interaction in the inherent mutagenicity of (+)-and (-)-BP 4,5-oxide in Chinese hamster V79 cells. These cells have very low or nondetectable levels of the mixed function oxidase system and epoxide hydrolase which are involved in the detoxification of BP 45oxide.